A phase interpolator is used to generate on its output fine phase steps or increments between two input signals (A and B). Typically the input signals A and B are cyclic, e.g. they are oscillator or clock(s).
The phase interpolator outputs a clock signal whose phase is in small increments between the inputs A and B as well as being bounded between them. Control signals are used to program the incremental phase of the outputs.
The phase interpolator's output is initially (minimum input control setting of all zeros) proportional to the A input signal. As the control signal is incremented, the output phase of the phase rotator is shifted from proportionally to the A input only, towards proportionally to the B input.
As the control input is changed to still larger values, the phase rotator output continues to shift the phase of the output and will eventually reach a point where it is solely proportional to the B input only. In this fashion the phase rotator output produces small increments in phase proportional and between its inputs A & B. The function is well known and used in many applications, such as in data communications between computer chips.